Live data from Hacker News

Viewing profile — nnx

nnx

HN member
Joined
Tue, Dec 17, 2013, 12:02 PM UTC
HN karma
6,153
Public activity
854 items

About nnx

CTO on a budget.

Recent public activity

  1. story
  2. story
  3. comment
    Comment #49093755

    is flickering fully fixed yet?

  4. story
  5. story
  6. story
  7. comment
    Comment #48512848

    Not saying there's no negative impacts, but what are the _huge_ negative impacts that have materialized so far?

  8. story
  9. story
  10. story
  11. story
  12. story
  13. comment
    Comment #47861683

    Coincidentally this is how pretraining works :)

  14. comment
    Comment #47824240

    Looks really interesting but the Go binding sadly uses cgo. Could the binding be done in pure Go? Or at least purego (the cgo alternative using Go assembly for FFI) ?

  15. story
  16. comment
    Comment #47789394

    I ended up just using the web version, which is actually better than the native app (multiple tabs works!).

  17. comment
    Comment #47789287

    > My M5 Pro can generate 130 tok/s (4 streams) on Gemma 4 26B. This seems high. At which quantization? Using LM Studio or something else? Note: Darkbloom seems to run everything on…

  18. story
  19. comment
    Comment #47610822

    The only limit is yourself!

  20. story
  21. story
  22. story
  23. story
  24. story
  25. comment
    Comment #47359913

    Are you `Ionstream` on OpenRouter? If so, it would be great to provide more models through OpenRouter. This looks interesting but not enough to make me go through the trouble of se…